INDEX

Explanations

sequences of character placeholders or formatting tags within a document

oai_token-act-pair · gpt-4o-mini Triggered by @bot

code structure and programming constructs

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

بوابة

-0.49

 незавершена

-0.48

uns

-0.41

 hotel

-0.40

CS

-0.39

 неу

-0.38

تج

-0.38

 Burns

-0.38

CWE

-0.37

MST

-0.37

POSITIVE LOGITS

 ब्रेकडाउन

0.55

 AssemblyTitle

0.54

olingo

0.51

OGND

0.51

曖昧さ回避

0.48

 WaitForSeconds

0.48

 noastră

0.47

majánló

0.47

verständlich

0.46

 üzere

0.46

Activations Density 0.039%

sequences of character placeholders or formatting tags within a document

code structure and programming constructs

No Comments

No Known Activations